PROJECT SUMMARY ! Major Depressive Disorder (MDD) and Generalized Anxiety Disorder (GAD) are debilitating and widespread disorders for which current treatments are unfortunately insufficient, particularly for associated cognitive symptoms. These cognitive symptoms, which include impairments of memory, executive function, and cognition, precede and predict the development of affective symptoms, thus they likely contribute to the development of associated affective symptoms. Given this, treating or preventing cognitive symptoms may prevent or lessen other symptoms of MDD and GAD. Unfortunately, our understanding of the neurobiological mechanisms underlying these cognitive symptoms is limited. A reliable and common cognitive deficit seen in MDD and GAD is overgeneralization of negative memories, defined as making sweeping conclusions based on a single past negative experience (e.g.: if I fail at this job interview, I will perform poorly on all my future interviews as well). Stress is known to promote generalization of negative memories; however, how this process occurs is not known. Thus, a better understanding of how stress alters neural circuits to promote generalization of negative memories is needed to elucidate the neural mechanisms underlying disease-relevant behaviors. On this basis, and consistent with NIMH's research domain criteria (RDoC) initiative, I propose to study the mechanisms of stress- induced generalization of negative memories (SIGNM) as an endophenotype of MDD and GAD. To this end, I developed an animal model for SIGNM by combining contextual fear conditioning with social defeat stress. I will use this model to elucidate cellular and molecular mechanisms of SIGNM. A promising candidate region is the dorsal hippocampus (DH) given (1) its role in associative learning, social memory, and episodic memory, (2) the fact that abnormal DH function has been implicated in generalization of memories in rodents and humans, and (3) that my pilot data shows that glutamatergic DH neurons mediate SIGNM. The DH is thought to encode contextual memories through organized ensembles of active neurons, commonly termed engram cells. Increased overlap of engram cells encoding different contextual memories is theorized to underlie increased generalizability between those contextual memories. Given this, increased overlap of neurons encoding negative event-associated and neutral contexts may provide a potential mechanism for SIGNM. I will explore this hypothesis in aim 1. To then explore circuit mechanisms of SIGNM, I will explore the contribution of DH glutamatergic outputs to RSC in SIGNM (Aim 2.) These projections are of promise as DH glutamatergic outputs to the RSC have been shown to mediate contextual fear memory processing. Next, as my pilot data shows that muscarinic acetylcholine receptors in the DH are necessary for SIGNM, I will then explore the role of cholinergic medial septum projections to the DH, the predominant source of DH acetylcholine, in SIGNM (Aim 3.) Upon completion of these experiments, I expect to have identified novel cellular and circuit mechanisms of SIGNM that can serve as RDoC for GAD and MDD, as well as novel therapeutic targets for these disorders.